FLOWFILL AND MSE BRIDGE APPROACHES: PERFORMANCE, COST, AND RECOMMENDATIONS FOR IMPROVEMENTS

Transcription

1 Report No. CDOT-DTD-R Final Report FLOWFILL AND MSE BRIDGE APPROACHES: PERFORMANCE, COST, AND RECOMMENDATIONS FOR IMPROVEMENTS Naser Abu-Hejleh Dennis Hanneman David J. White Trever Wang Ilyess Ksouri February 2006 COLORADO DEPARTMENT OF TRANSPORTATION RESEARCH BRANCH

2 The contents of this report reflect the views of the author(s), who is (are) responsible for the facts and accuracy of the data presented herein. The contents do not necessarily reflect the official views of the Colorado Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. Use of the information contained in the report is at the sole discretion of the designer. i

3 Technical Report Documentation Page 1. Report No. CDOT-DTD-R Government Accession No. 3. Recipient's Catalog No. 4. Title and Subtitle FLOWFILL AND MSE BRIDGE APPROACHES: PERFORMANCE, COST, AND RECOMMENDATIONS FOR IMPROVEMENTS 5. Report Date February Performing Organization Code 7. Author(s) Naser Abu-Hejleh, Dennis Hanneman, David J. White,Trever Wang, and Ilyess Ksouri 9. Performing Organization Name and Address Colorado Department of Transportation 4201 E. Arkansas Ave Denver, Colorado Performing Organization Report No. CDOT-DTD-R Work Unit No. (TRAIS) 11. Contract or Grant No. 12. Sponsoring Agency Name and Address Colorado Department of Transportation - Research 4201 E. Arkansas Ave. Denver, CO Type of Report and Period Covered 14. Sponsoring Agency Code Study # Supplementary Notes Prepared in cooperation with the US Department of Transportation, Federal Highway Administration 16. Abstract: Construction of a typical Colorado DOT (CDOT) bridge approach structure includes placement of a high quality backfill material behind the abutment wall, and installation of a concrete approach slab supported by the bridge abutment wall at one end and the sleeper slab foundation at the roadway end. Since 1992, three new alternatives for the abutment backfill have been employed by CDOT: (1) relatively expensive flowfill; (2) lower cost mechanically stabilized earth (MSE) using granular, well-graded Class-1 Backfill, and (3) MSE using free draining Class B Filter soil. However, bridge bump problems at the sleeper slab are still occurring. In the Founders/Meadows bridge structure, both the bridge footings and approaches are supported by geosyntheticreinforced soil (GRS) walls to minimize the uneven settlements between the bridge and its approaches (called GRS Abutment ). Since this structure is unique, performance data from gauges embedded in the approaches and from smoothness tests were collected over five years. The objective of this study is to improve CDOT s current practice for bridge approaches (improve performance and reduce costs) based on the following information obtained in this study: 1) comments and suggestions collected from CDOT Staff and reported in the literature; 2) performance and cost-effectiveness of CDOT s MSE and flowfill bridge approaches and performance and design assessment of the Founders/Meadows GRS approaches; and 3) causes and sources of the bridge approach settlement problems observed in some of CDOT s MSE and flowfill bridge approaches. Implementation Statement: CDOT should use the lower cost MSE approaches with either Class B or Class 1 Backfill materials in its future projects. Flowfill should remain a viable alternative for certain field and construction scenarios that justify its higher costs. The use of an MSE or GRS abutment system should be considered as a viable alternative for all future bridges - it is the best system to alleviate the bridge bump problem. Additional recommendations are furnished to improve CDOT current practice and mitigate the bridge approach settlement problem in Colorado s new bridge approaches: 1) approach slab and roadway: select the appropriate construction elevation grades, and length of approach slab based on the projected post-construction settlements, warranty and smoothness requirements; 2) better support and drainage systems for the sleeper slab where the settlement problem occurs; 3) new and tightened construction requirements for the flowfill, Classes 1 and B Backfill materials; 4) suggestions for construction of tiered MSE wall around the bridge approaches; 5) suggestions for subsurface geotechnical investigation and settlement analysis of the fill and foundation soil layers; 6) design parameters for MSE approaches; and 7) recommendations for the bridge expansion devices and drainage measures. Finally, recommendations for the forensic investigation and repair of existing bridge approaches are presented. 17. Keywords bumps, approach slabs, abutments, backfill, cost-effectiveness, settlement, sleeper slabs, drainage, construction, repair, inspection, foundations, embankments, expansion devices, walls 18. Distribution Statement No restrictions. This document is available to the public through the National Technical Information Service, Springfield, VA Security Classif. (of this report) Unclassified Form DOT F (8-72) 20. Security Classif. (of this page) Unclassified Reproduction of completed page authorized 21. No. of Pages Price ii

4 CONVERSION TABLE U. S. Customary System to SI to U. S. Customary System (multipliers are approximate) Multiply To Get Multiply by To Get (symbol) by (symbol) LENGTH Inches (in) 25.4 millimeters (mm) mm in Feet (ft) meters (m) m 3.28 ft yards (yd) meters (m) m 1.09 yd miles (mi) 1.61 kilometers (km) m mi AREA square inches (in 2 ) square millimeters (mm 2 ) mm in 2 square feet (ft 2 ) square meters (m 2 ) m ft 2 square yards (yd 2 ) square meters (m 2 ) m yd 2 acres (ac) hectares (ha) ha 2.47 ac square miles (mi 2 ) 2.59 square kilometers (km 2 ) km mi 2 VOLUME fluid ounces (fl oz) milliliters (ml) ml fl oz gallons (gal) liters (l) l gal cubic feet (ft 3 ) cubic meters (m 3 ) m ft 3 cubic yards (yd 3 ) cubic meters (m 3 ) m yd 3 MASS ounces (oz) grams (g) g oz pounds (lb) kilograms (kg) kg lb short tons (T) megagrams (Mg) Mg T TEMPERATURE (EXACT) Farenheit ( F) 5(F-32)/9 Celcius ( C) C 1.8C+32 F (F-32)/1.8 ILLUMINATION foot candles (fc) lux (lx) lx fc foot-lamberts (fl) candela/m (cd/m) cd/m fl FORCE AND PRESSURE OR STRESS poundforce (lbf) 4.45 newtons (N) N.225 lbf poundforce (psi) 6.89 kilopascals (kpa) kpa.0145 psi iii

5 FLOWFILL AND MSE BRIDGE APPROACHES: PERFORMANCE, COST, AND RECOMMENDATIONS FOR IMPROVEMENTS by Naser Abu-Hejleh, Ph.D., P.E., FHWA Resource Center Dennis Hanneman, P.E., Kleinfelder, Inc. David J. White, Ph.D., Iowa State University Trever Wang, Ph.D., P.E., Colorado DOT (Bridge) Ilyess Ksouri, Colorado DOT (Geotechnical Program) Report No. CDOT-DTD-R Sponsored by the Colorado Department of Transportation In Cooperation with the U.S. Department of Transportation Federal Highway Administration February 2006 Colorado Department of Transportation Research Branch 4201 E. Arkansas Ave. Denver, CO (303) iv

6 ACKNOWLEDGEMENTS The Colorado Department of Transportation and the Federal Highway Administration provided funding for this study. Thanks are extended to CDOT Maintenance Offices in Regions 1, 2, 4 and 6 who helped with collecting inspection records of the bridge approaches and to Steve White from CDOT Staff Bridge who provided valuable information on all bridges presented in this study. The CDOT Study Panel on this study and others provided in-depth technical review of this report and valuable comments from: Prof. George Hearn (University of Colorado at Boulder), Prof. Jorge G. Zornberg (University of Texas at Austin), Matt Greer (Colorado FHWA Division Office), Rich Griffin, Rene Valdez, Mike McMullen, Trever Wang, Cheng Su, Dennis Rhodes, Rene Valdez, Mike Gdovin, Aziz Khan, Skip Outcalt, Dean Sandoval, Tom Wrona, Corey Stewart, Mike Day, Pete Graham, Dick Osmun, and Alan Hotchkiss. Their knowledge and advice kindly offered in meetings, s, telephone conversations, and field visits were essential to the successful completion of this report. Substantial help and support to this research were provided by Rich Griffin. Thank you all. v

7 EXECUTIVE SUMMARY The current typical Colorado DOT (CDOT) bridge approach system includes a foundation soil layer, an embankment fill soil layer, a high quality backfill material placed behind the abutment wall, a concrete approach slab supported by the bridge abutment wall at one end and the sleeper slab foundation at the roadway end (the sleeper slab is placed on the abutment backfill), a drainage system, and an expansion joint. Settlement at the sleeper slab leads to an abrupt change in elevation grade: a bump. Since 1992, three new alternatives for the abutment backfill have been used by CDOT: (1) relatively expensive flowfill (a low-strength concrete mix, 110 bridges, ); (2) lower cost mechanically stabilized earth (MSE) (14 bridges, ) with granular and well-graded Class-1 Backfill soil; and (3) MSE system with free draining Class B Filter soil (10 bridges, 2002 to 2005). Despite some performance improvements with these alternatives, the occurrence of significant approach settlement problems in flowfill and MSE approaches, resulting in high repair costs, is still being reported. In the Founders/Meadows bridge structure, both the bridge footings and approaches are supported by geosyntheticreinforced soil (GRS) walls to minimize the uneven settlements between the bridge and its approaches (called GRS Abutment ). Performance data from instrumentation embedded in the approaches and smoothness tests were collected periodically over five years. The primary objective of this study is improve CDOT s current practice for bridge approaches (improve performance and reduce costs) from results of the following tasks: 1) Document CDOT s current practice for the geotechnical investigation, construction, and repair of bridge approaches and the comments and suggestions collected from CDOT Staff and reported in the literature to improve this practice; 2) Develop and apply a forensic investigation to determine the causes, sources, and time progression of the settlement problems experienced in CDOT s MSE and flowfill bridge approaches; 3) Evaluate the performance of CDOT s MSE and flowfill approaches and performance and design assessment of the Founders/Meadows bridge approaches; and 4) Estimate the total unit cost (construction and repair) needed to maintain acceptable performance of CDOT s flowfill and MSE bridge approaches over their entire service life (for comparison between flowfill and MSE approaches). vi

8 Approach Settlement Problem in Colorado. There are four main causes of the observed bridge bump problems. First, the elevation grades of the as-built bridge and roadway approaches do not exactly match the design elevations leading to creation of a bump at the end of construction. The problem is worsened if the expansion joint is placed per the design elevation and not based on the as-built grades of the constructed bridge and approaching roadways. Second, failure of the installed drainage measures to keep surface and excess ground water from reaching the fill and foundation soil layers, which is a common factor in almost all the bridge approaches that experienced settlement problems. Water contributes to softening in soil zones between the granular soil layer and the underlying fine-grained soil layer. Third, settlement of the placed fill materials during or shortly after construction is completed can be due to lack of adequate compaction, construction during the cold season with frozen fill, and placement of fill materials dry of optimum leading to compression of the soil following subsequent wetting. And fourth, settlement of the compressible clay foundation soil layer that may not be detected or adequately addressed during design from the available subsurface investigation information. For example, water can soften the top of a clay foundation soil layer that derives apparent strength from desiccation. Performance and Cost Results. Most of the flowfill and MSE bridge approaches constructed by CDOT since 1993 are performing well, with no settlement or cracking problems. Most of the settlement problems for the flowfill approaches are associated with the older bridge approaches constructed before 1994 when CDOT just started using flowfill. Out of 28 bridge approaches constructed with MSE Class-1 Backfill, 4 approaches failed due to poor construction operations. Performance/cost analyses indicate that the use of MSE Class 1 Backfill is more cost-effective than flowfill only if the rate of repair of MSE approaches will decline in the future. No problems are reported for the Class B Backfill approaches. The overall short- and long-term performance of the GRS approaches of the Founders/Meadows structure is excellent. Temperature has a significant effect on integral abutments, leading to continuous cyclic lateral movements of the MSE backfill with time (compression and expansion movement) and to cyclic lateral earth pressures (passive and active) against the abutment wall. Continuous expansion of the MSE backfill of approximately 1.5 to 2 mm every year was noticed. The presence of compressible polystyrene sheets behind abutment walls accommodated to a large extent (but not entirely) the vii

9 thermal expansion movement of the bridge superstructure and reduced the active lateral earth pressure to almost zero. Implementation Statement: It is recommended that CDOT use the lower cost MSE approaches with either Class B or Class 1 Backfill materials in its future projects over the next few years and monitor their performance and document their repair costs. Flowfill should remain a viable alternative for certain field and construction scenarios that justify its higher costs. Warranty and smoothness requirements for bridge approaches are presented along with recommendations for construction of a tiered MSE wall system around the bridge approaches. Two new supporting systems for the sleeper slab are suggested: the first system consists of placing most of the high quality MSE backfill under the sleeper slab rather than the approach slab (as currently employed by CDOT). The second supporting system consists of using driven piles to support the sleeper slab and using the much cheaper Class 2 Backfill material behind the abutments. The length of approach slab (L) should be related to the projected long-term settlement (Δ) of the sleeper slab that would occur after the pavement structure is placed such that Δ/L < The study provides examples for computing the settlement of fill and foundation soil layers. Replacement of the concrete approach and sleeper slabs with full depth asphalt approach slabs should be considered when the settlement is significant and occurring for a long period of time. Regular maintenance overlays will be needed. The expansion device placed on top of the sleeper slab should be installed at an elevation that matches the as-constructed and surveyed grades of the bridge and approach roadway (minor adjustments from the design elevations), or even higher by up to one inch (for approach slab of 20 ft long) to compensate for the anticipated post-construction settlements (if pre-loading is not performed). viii

10 With regard to placement of backfill materials, it is suggested to add a construction requirement of vibration for the flowfill and compaction requirements as those established by CDOT for rocky embankments for Class B Filter soil. Further, it is recommended that CDOT s specifications for compaction of abutment and embankment fill soils be more rigorous, especially during the cold season, and for compaction of the top of a desiccated foundation soil layer that is susceptible to wetting induced softening. Compact granular fill soils wet of the optimum moisture and, after compaction is completed, consider dousing the soil with water to reduce the potential for future collapse. For MSE approaches, the recommended design active and passive earth pressures should be considered with caution because they are based on limited data. It is recommended to use a softer (less dense) and thicker compressible (e.g., polystyrene) sheet in the upper zone of the abutment wall to minimize lateral earth pressures. This study provides some criteria for selecting the appropriate location of test holes in the subsurface geotechnical investigation and suggests application of seasonal corrections to the measured SPT data if they are collected during the dry or cold seasons of the year. Three new measures are recommended to ensure that the joint placed above the sleeper slab does not allow water to seep into the soil under the sleeper slab. In addition it is recommended that drainage inlets at the end of a bridge deck to collect surface water before getting to the approach slab be adopted as standard design detail on all bridges. Current problems with drainage pipes, which seem not to work in many cases, should be corrected. It is recommended to place the expansion device over the abutment wall and not over the sleeper slab to prevent dragging of the approach slab which results in cracking. Finally, recommendations for forensic investigations and repair of bridge approaches are outlined. ix

11 TABLE OF CONTENTS 1.0 INTRODUCTION Background CDOT s Needs for Bridge Approaches Study Objectives and Overview of the Report CDOT'S CURRENT PRACTICE FOR BRIDGE APPROACHES Overview Abutment Walls, Foundations, and Wing Walls Concrete Approach Slab Abutment Backfill Conventional Granular Class 1 Structure Backfill Flowfill Structural Backfill Mechanically Stabilized Class-1 Backfill Mechanically Stabilized Class-B Filter Material GRS Abutment System Approach Embankment Bridge/Roadway Drainage System Bridge Deck Expansion Foundation Investigation at Bridge Approaches Subsurface Geotechnical Investigation Settlement Analysis Soil Improvement Techniques Repair of Colorado Bridge Approaches Overview Soil Stabilization Techniques Additional Recommendations from CDOT Staff to Improve Current Practice for Construction of Bridge Approaches BRIDGE APPROACH SETTLEMENT PROBLEM: CAUSES AND FORENSIC INVESTIGATION Overview Causes of the Soil Settlement at the Sleeper Slab x

12 3.2.1 Influence of Moisture and Temperature Changes The Forensic Investigation: Needed Information Design, Materials, and Construction Records of the Bridge Approach Structure Level, Location, and Time Progress of the Approach Settlement Problem Subsurface Geotechnical Investigation PERFORMANCE AND COST-EFFECTIVENESS ANALYSIS OF CDOT FLOWFILL AND MSE BRIDGE APPROACHES Overview Performance of Side by Side Bridge Flowfill and MSE Bridge Approaches Procedure for Evaluation of Performance and Cost-Effectiveness of Flowfill and MSE Bridge Approaches Performance from Records of CDOT's Bridge Management Section Performance Based on Bridge Approach Settlement Information on Applied and Required Repair Measures Cost-Effectiveness Analysis of Flowfill and MSE Bridge Approaches Region 6 Bridge Approaches Region 4 Bridge Approaches Region 1 Bridge Approaches Region 2 Bridge Approaches Performance and Cost-Effectiveness of Flowfill and MSE Bridge Approaches INVESTIGATION OF BRIDGES WITH SEVERE BRIDGE APPROACH SETTLEMENT PROBLEMS Objectives Salt Creek (L-18-BD) (MSE Abutment Backfill) Description of the Bridge Structure Description of the Bridge Bump Problem Review of the Construction Plans and Geotechnical Report Results of the Subsurface Geotechnical Investigation Backfill and Embankment Materials Foundation Soils Settlement Analysis and Results xi

13 5.2.6 Causes of the Approach Settlement Problem Sources, Magnitude, and Timing of the Approach Settlement Problem I-70/I-225 Interchange (Flowfill Abutment Backfill) Overview of the Problem and Construction Plans Results of Surface Inspection Results of the Subsurface Geotechnical Investigation Settlement Analysis and Results Concluding Remarks and Recommendations SH 287 Over Little Thompson River (C-16-DK) (Flowfill Abutment Backfill) Overview Results of the Subsurface Geotechnical Investigation Fill Materials Foundation Soils Settlement Analysis of the Sleeper Slab Concluding Remarks Structure E-19-Z on US 36 East of Bennett (MSE Abutment Backfill) Site Conditions and Observations of the Problem Subsurface Geotechnical Investigation Causes of the Approach Settlement Problem Repair of Bridge Approaches Structure E-17-PR at I-76 at 136 th Ave (Flowfill Abutment Backfill) Site Conditions Subsurface Geotechnical Investigation Causes of the Bridge Approach Settlement Problem Recommendations PERFORMANCE OF THE FOUNDERS/MEADOWS MSE BRIDGE APPROACHES OVER 5 YEARS Introduction Overview of the Study Investigation Overall Performance of the Bridge Approaches Moisture Changes in the MSE Abutment Backfill xii

14 6.5 Lateral Strains and Earth Pressures in the Abutment MSE Backfill Measured Strains in the Reinforcements of the MSE Abutment Backfill Horizontal Soil Pressure against the Abutment Back Wall Design Implications of the Measured Results SUMMARY, STUDY FINDINGS AND RECOMMENDATIONS Overview Study Findings CDOT's Current Practice for Bridge Approaches and Reported Suggestions For Improvement Causes of the Bridge Approach Settlement Problem in Colorado Performance of Different Systems Employed for Construction of Bridge Approaches Recommendations Systems for Construction of Bridge Approaches General Recommendations Better Support and Drainage Systems for the Sleeper Slab Approach Slab Flowfill Abutment Backfill MSE with Class B Filter Material MSE Backfill & Embankment Influence of Temperature Changes on Integral Abutments Foundation Investigation at Bridge Approaches Bridge Expansion Device Drainage Measures Repair of Colorado Bridge Approaches REFERENCES APPENDIX A: CONSTRUCTION DETAILS OF CDOT'S BRIDGE APPROACHES. A-1 APPENDIX B: DETAILED PERFORMANCE DATA OF CDOT'S FLOWFILL AND MSE BRIDGE APPROACHES... B-1 xiii

15 APPENDIX C: PHOTOS OF VARIOUS FLOWFILL AND MSE BRIDGE APPROACHES IN COLORADO... C-1 xiv

16 LIST OF FIGURES Figure 4.1 Measured Elevation Profiles for Structure C-15-U Where Side by Side Flowfill and MSE Approaches Were Constructed Figure 4.2 Use of Caissons in the Repair of the Bridge Approach Settlement Problem at Structure E-17-PQ Figure 5.1 A Section along the Front MSE Wall of the Salt Creek Bridge Figure 5.2 A Section along the Upper and Lower MSE Walls of the Salt Creek Bridge Figure 5.3 An Elevation Profile of the Bridge Approach along the Shoulder Line of the SW Corner of the Salt Creek Bridge Approaches Figure 5.4 A Settlement Profile of the Bridge Approach along the Shoulder Line of the SW Corner of the Salt Creek Bridge Approaches Figure 5.5 Elevation Profiles across the SW Corners of the Old (dashed line) and New (solid line) Bridge Approaches of the Salt Creek Bridge Figure 5.6 Boring Logs for the Salt Creek Bridge Approaches Figure 5.7 Influence of Wetting on a Remolded Soil Sample of Loosely Compacted Class1 Backfill (Salt Creek Bridge Approaches) Figure 5.8 Influence of Wetting on a Remolded Soil Sample of Well-Compacted Class1 Backfill (Salt Creek Bridge Approaches) Figure 5.9 Predicted Settlements of the Sleeper Slab Due to Consolidation of the Foundation Clay Layer (Salt Creek Bridge Approaches) Figure 5.10 Details of the Eastern Abutment of the WB I-70 to SB I-225 Bridge Figure 5.11 Details of the CIP Retaining Wall Constructed along the Northern-Eastern Side of the I-70/I-225 Ramp Figure 5.12 Elevation Profiles of the Eastern Approaches to the I-70/I-225 Structure Figure 5.13 Boring Log for the I70/I225 Bridge Structure Figure 5.14 Predicted Sleeper Slab Consolidation Settlement of the I-70/I-225 Bridge Approaches Figure 5.15 Boring Log for the SH 287 Structure Figure 5.16 Predicted Settlement with Time at the Sleeper Slab of the SH 287 Bridge Figure 5.17 Elevation Profile of the Approaches of Structure E-19-Z Before Repair xv

17 Figure 6.1 Typical Section through Front and Abutment GRS Walls of the Founders/Meadows Bridge Structure Figure 6.2 Instrumentation Layout of Section 800 of the Founders/Meadows Bridge Approaches Figure 6.3 Locations of the Profiles Lines in the Founders/Meadows Bridge Structure Figure 6.4 Measured Elevation Profiles Relative to the Bridge Abutment for Two Lines over the East Abutment along the Traffic Direction Figure 6.5 Measured Elevation Profiles Relative to the Bridge Abutment for Two Lines over the West Abutment along the Traffic Direction Figure 6.6 Measured Changes in Soil Moisture Before and After Placement of the Drainage Protection System Figure 6.7 Measured Air Temperature below Girders of the Founders/Meadows Structure Figure 6.8 Measured Geogrid Tensile Strains at Various Times from Gages Placed in the MSE Abutment Backfill Behind the Abutment Wall as Shown in Figure Figure 6.9 Rough Estimate of the Expansion Lateral Movements of the MSE Abutment Backfill during Various Times Figure 6.10 Measured Lateral Earth Pressure Against the Lower Portion of the Abutment Wall of the Founders/Meadows Bridge Figure 6.11 Measured Lateral Earth Pressure Against the Top Portion of the Abutment Wall of the Founders/Meadows Bridge (see Figure 6.2) during Various Times Figure 7.1 The Driven Pile System Proposed for the 120 th Project Bridge Approaches Figure 7.2 The MSE Wall System under the Sleeper Slab Proposed for the 120 th Project Bridge Approaches Figure 7.3 The Most Economical Form of Drainage System Recommended for the Sleeper Slab xvi

18 LIST OF TABLES Table 2.1 CDOT Material and Construction Requirements for the Granular Class1 Backfill 2-4 Table 2.2 CDOT Material Requirements for Flowfill Backfill Table 2.3 CDOT Specifications for Class B Filter Materials Table 2.4 CDOT Description of Soils Based on SPT-N Values Table 2.5 Typical CDOT Settlement Recommendations for Problematic Foundations Soils Table 4.1 Ten Bridge Structures with Best Flowfill Approaches in Region Table 4.2 Ten Bridge Structures with Worst Flowfill Approaches in Region Table 4.3 Region 4 Problematic Flowfill Bridge Approaches Table 4.4 Problematic Bridge Approaches in Region Table 4.5 Performance of Flowfill and MSE Bridge Approaches Table 4.6 Cost-Effectiveness Analysis of Flowfill and MSE Bridge Approaches Table 5.1 Laboratory and Field Test Results for the Fill Material of the Salt Creek Bridge Approaches Table 5.2 Laboratory and Field Test Results for the Foundation Soil of the Salt Creek Bridge Approaches Table 5.3 Laboratory and Field Test Results for the Embankment Fill of the I-70/I-225 Interchange Approaches Table 5.4 Laboratory and Field Test Results for the Embankment Material at the SH287 Bridge Table 5.5 Laboratory and Field Test Results for the Foundation Soil of the SH287 Bridge 5-42 Table 6.1 Estimated Active, At Rest, and Passive Abutment Lateral Earth Pressures xvii

19 1. INTRODUCTION 1. Background A bump often develops at the end of a bridge near the interface between the abutment and the approaches. The main cause of uneven settlements in typical bridge foundation systems is the use of different foundation types. That is, while the approaching roadway structure is typically founded on compacted backfill soil, the bridge abutment is typically founded on much stronger soils or bedrock by deep foundations. Bridge bumps cause uncomfortable rides, create hazardous driving conditions, and require costly, frequent repairs with traffic delays. The problem affects 25% of the bridges in the United States, approximately 150,000 bridges, and the amount of maintenance required is estimated to be at least $100 million every year (NCHRP Synthesis 234, 1997). The bump problem is a complex problem involving a number of components, including the natural foundation soil, the fill material, the foundation type used to support the abutment, the abutment type, the structure type, the bridge/roadway joints, the approach slab, roadway, and the construction methods. Numerous investigations have been undertaken during the past decades to identify the causes and minimize the differential settlements between the bridge abutments and their approaches. The most commonly reported causes of the bump in order of importance (NCHRP Synthesis 234, 1997) are: 1. Compression of the fill material 2. Settlement of the foundation soil 3. Poor construction practice 4. Poor drainage 5. Poor fill material 6. Loss of fill by erosion 7. Poor joints 8. Temperature cycles 1.2 CDOT s Needs for Bridge Approaches Before 1992, the Colorado Department of Transportation s (CDOT s) measures to alleviate the bridge bump problem included the extension of wing walls along the roadway shoulders, and use of approach slabs and granular backfill (Class I Structural Backfill) behind the abutments. The approach slab is supported by the bridge abutment wall at one end and a sleeper slab foundation at the roadway end. The use of an approach slab allows a gradual distribution of any approach settlement and a smoother transition between the bridge and approaching roadway. CDOT began 1-1

20 using concrete approach slabs in the 1970s, and the majority of CDOT s bridge approaches built since 1990 have been constructed with the approach slab tied to the abutment wall. Since 1992, CDOT has implemented several additional improvements to construction of bridge approaches in order to alleviate the bridge bump problem. Most of CDOT s new bridges are constructed with an integral abutment system where the abutment and the superstructure are rigidly connected to eliminate or reduce joints in the bridge superstructures. The integral end diaphragm type abutments are mostly supported by deep foundations. It is estimated that 10% of CDOT s bridge abutments are supported by shallow foundations, 50% by drilled shafts, and 40% by driven piles. Temperature cycles are more critical in integral abutments since the expansion and contraction of the bridge decks and girders lead to lateral displacement of the approach backfill. To overcome this problem, a very small gap or a compressible material (around 15 cm in thickness) is incorporated between the abutment fill and the bridge abutment. Four new systems for construction of bridge approaches have been implemented: A. Flowfill Bridge Approaches. In November 1992, CDOT began using flowfill (a low-strength concrete mix) backfill behind the abutment wall to reduce the approach settlements. The selfleveling ability of flowfill allows it to flow, so no compaction is needed, and fill voids and hardto-reach- zones (curved and cornered zones). Also, it experiences negligible settlements after curing. A total of 110 bridges were constructed with flowfill abutment backfill from 1993 to 2001 (none could be found in 2002 and 2003). Given the high cost of flowfill ($76 per cubic yard in 2005) relative to conventional embankment material, the performance of existing installations should be studied to determine if this practice is worth continuing. B. MSE Class 1 Backfill Bridge Approaches. The use of MSE (mechanically stabilized earth) Class-1 backfill behind abutments wall as a lower cost alternative to flowfill ($37 /CY in 2005) has been a growing practice in Colorado. Standard details for MSE abutment Class-1 backfill were introduced in CDOT on May 21, A total of 14 bridges were constructed with MSE Class 1 Backfill between 1999 and 2003 (none before 1999). Most of the reinforcements in the MSE embankments are geofabric wrapped around the back face of the abutment, but geogrid (stiffer) reinforcements are considered in some situations to stiffen the backfill and further 1-2

21 reduce the approach settlements. The reinforced fill behind the abutment is used to build a vertical, self-contained wall capable of holding an approximately vertical shape and forming an air gap between the abutment and retained fill. By installing tensile reinforcements in the fill, it has been reported that a stiffer approach would be created. However, no field performance records are available for CDOT s MSE approaches with Class-1 Backfill. C. MSE Class B (Porous) Backfill Bridge Approaches. In the last few years, Class B filter material has replaced the Class-1 Backfill in construction of 10 new MSE backfill bridges (cost $57/CY in 2005). Class B filter material was selected because it is more free draining, is less susceptible to wetting induced softening/collapse, less erodible, has less fines for clogging drainage systems, and requires less compaction effort compared to Class-1 Backfill. Although these engineering properties are superior, performance information of the system was still needed. Note that in Systems A, B, and C, the abutment walls were supported by deep foundations and this is not the case with System D presented next. D. GRS Abutment System. In the Founders/Meadows bridge structure, constructed in 1999 near Denver, Colorado, geosynthetic-reinforced soil (GRS) walls were employed to support the shallow footings of a two-span bridge and the approaching roadway structures (see Chapter 6 for complete details, Abu-Hejleh et. al., 2000 and 2001 for references). The approaching roadway embankment and the bridge footing were integrated at the Founders/Meadows structure with an extended reinforced soil zone in order to minimize/alleviate the uneven settlements between the bridge abutment and approaching roadway (main cause of the bridge bump problem). This structure was considered experimental, and its approaches were instrumented during construction with moisture gages, strain gages, and pressure cells and profilometer tests were conducted to evaluate the smoothness of the approaches. Monitored performance data were periodically collected from beginning of construction through five years of service. There is a need to document and evaluate these data and summarize the lessons that are learned from this unique structure. 1-3

22 1.3 Study Objectives and Overview of the Report But significant settlement at the sleeper slab in flowfill and MSE approaches still occurs, causing an abrupt change in elevation grade at the sleeper slab, resulting in high repair costs. Performance information on the recent systems and measures employed by CDOT to alleviate the bridge bump problem are needed. CDOT engineers need to know if their current practice is worth continuing or how it can be improved. In particular, the causes and sources of the approach settlement problem in Colorado bridges must be identified so that best practices can be implemented. This study was proposed to address all these needs. The objective of this study is to provide recommendations to improve CDOT s current practice for construction of bridge approaches (improve performance and reduce costs). Several tasks were performed to meet this objective: 1. Summarize CDOT s current practice that has evolved since 1993 for the geotechnical investigation, construction, and repair of bridge approaches and the comments and suggestions collected from CDOT Staff and reported in the literature to improve this practice (Chapter 2 and Appendix A). The current typical CDOT bridge approach system includes a foundation soil layer where subsurface geotechnical investigation is performed followed by performing settlement analysis, an embankment fill soil layer placed on top of the foundation soil layer, a high quality backfill material (flowfill or MSE Backfill) placed behind the abutment wall and beneath the approach and sleeper slabs (described before), surface and internal drainage systems, and expansion joint device typically placed on top of the sleeper slab. 2. Provide detailed descriptions of all possible causes of the bridge approach settlement problem at the sleeper slab and the information needed in a forensic investigation to identify the causes and sources of this problem and determine if the settlement problem has more or less ended or if significant settlement potential remains in the future (Chapter 3). The causes include compression and creep movements of the fill and foundation soil materials (due to compressible soil layers, and applied static and dynamic loads), thermal movements of the bridge superstructure (of more concern with integral abutments), lateral movement of side walls (MSE 1-4

23 walls must laterally move to mobilize the tensile resistance of its reinforcements) problems in the geotechnical investigation, problems encountered during construction, and inadequate performance of the expansion joints and the drainage systems. A detailed description of the influence of moisture and temperature on soil settlements is presented. The information needed in a forensic investigation is: design, materials, and construction records of the bridge approach; structure, level, location, and time progression of the settlement problem; and information from a comprehensive subsurface geotechnical investigation that is described. Chapters 3, 4 and 5 present the procedure to collect and analyze this information for bridge approaches that experienced approach settlement problems. 3. Evaluate and compare the field performance and cost-effectiveness of bridge approaches constructed by CDOT with flowfill and MSE backfill materials (Chapter 4, Appendices B and C). Performance of side by side flowfill and MSE bridge approaches is presented for two bridges in Region 4. A procedure was developed and applied to: 1) evaluate the performance of MSE and flowfill bridge approaches, and 2) Estimate the total unit cost (construction and repair) needed to maintain acceptable performance of CDOT s flowfill and MSE bridge approaches over their entire service life (for comparison between flowfill and MSE approaches). The performance and cost (construction and repair costs) information were obtained from records collected by the CDOT Bridge Management Section, input from CDOT s Regional Maintenance Offices, and field visits, and from information published by the CDOT Engineering Estimates & Market Analysis Unit. Performance ratings for bridge approaches reflected the range of settlement experienced by the bridge approaches at the sleeper slab and the traffic speed (significant to moderate to slight bump problems). 4. Conduct a forensic investigation (including short- and long-term settlement analyses) on the MSE and flowfill bridge approaches that experienced significant settlement problems. The purpose is to determine the causes and sources of the current settlement problem and if this settlement has more or less ended or if significant settlement potential remains in the future (Chapter 5 and Appendix C). This information is needed to develop an effective plan for repair and mitigation of the settlement problem. The investigation was performed on five bridge structures, with three thoroughly investigated: 1-5

24 a) Salt Creek Bridge along SH 50 (L-18-BD, MSE Backfill) in Region 2. b) SH 287 Over Little Thompson River (C-16-DK, flowfill) in Region 4. c) I-70/I-225 Interchange in Region 6 (flowfill). And two more were previously investigated by CDOT Soil and Foundation Units: d) Structure E-19-Z on US 36 East of Bennett in Region 1 (MSE Backfill). e) Structure I-76 at 136 th Ave in Region 6 (flowfill). 5. Compile and analyze the data collected over five years on the performance and design assessment of the measures employed in the Founders/Meadows Bridge to alleviate the bridge bump problem (Chapter 6). Analysis of these performance data also provides insight into the behavior and validity of some of the design assumptions for MSE and Flowfill bridge approaches. 6. Based on the results of previous tasks summarize the study findings and the recommendations learned to improve CDOT s current practice for bridge approaches (Chapter 7). 1-6

25 2. CDOT S CURRENT PRACTICE FOR BRIDGE APPROACHES 2.1 Overview The evolution of CDOT s practice since 1993 for construction of bridge approaches is described in this chapter. The bridge approach system includes the following components: approach slab, drainage system and expansion joints, abutment backfill, embankment fill, and foundation soil. The main details of the bridge approach system are listed in Appendix A (Figures A.1 to A.6). Currently, CDOT has standards for details of MSE or flowfill abutment backfill materials used with concrete approach slabs (See Figures A.1 and A.2.) or without approach slabs (with asphalt approach slabs, see Bridge Worksheet B-206-M2 at Most (estimated 95%) of CDOT s bridge structures built in the last 13 years (since 1992) were constructed with a concrete approach slab (see Figures A.3, A.4, and A.5 for details) and a bridge expansion device (Figure A.6) installed at the sleeper slab foundation. The main sources for the information presented in this chapter are the CDOT Bridge Web Page that can be accessed online at and CDOT Standard Specifications that can be accessed online at Abutment Walls, Foundations, and Wing Walls Most of CDOT s new bridges are constructed with an integral abutment system where the abutment and the superstructure (girders and decks) are rigidly connected (through steel reinforcement and monolithic pour of concrete) to eliminate or reduce joints in the bridge superstructures. The integral, end diaphragm type abutments are mostly supported by foundations. It is estimated that 10% of CDOT s bridge abutments are supported by shallow foundations, 50% by drilled shafts, and 40% by driven piles. The abutments are usually supported on deep foundations because of stipulations such as bridge scour, and cost and benefit of carrying heavy loads with the latter being more dominant. Deep foundations are the most efficient means of transferring heavy loads from superstructures to substructures and bearing materials without significant distress from excessive settlement. 2-1

26 It has been a common practice at CDOT to use wingwalls for U-type Abutments. Normally, a wing wall will be cantilevered off the abutment. When the required wingwall length exceeds a practical length, a retaining wall is recommended. The same foundation system is recommended for supporting both the retaining and abutment walls to reduce risk of the retaining wall settlement relative to the abutment. The wing wall could be replaced with MSE walls, especially if MSE walls are employed to support the soil beneath the abutment wall. 2.3 Concrete Approach Slab Subsection 7.3 of the CDOT Bridge Design Manual reads as follows. Approach slabs are used to alleviate problems with settlement of the bridge approaches relative to the bridge deck. The main causes of this settlement are movement of the abutment, settlement and live load compaction of the backfill, moisture, and erosion. Approach slabs shall be used under the following conditions: 1. Overall structure length greater than 250 feet. 2. Adjacent roadway is concrete. 3. Where high fills may result in approach settlement. 4. When the District requests them. 5. All post-tensioned structures. CDOT started using concrete approach slabs in the 1970s. Most (estimated 95%) of CDOT s bridge structures built in the last 13 years (since 1992) were constructed with concrete approach slabs. Construction of a new approach concrete slab is also considered in the rehabilitation of old bridge approaches. The remaining 5% of bridge approaches are constructed with asphalt approaches. Use of asphalt approaches is considered with smaller span bridges, bridges with low traffic, and when an adjacent roadway is asphalt. CDOT provides three different details for the use of concrete approach slabs: 1. Concrete approach slab overlaid by 3 hot bituminous pavement (HBP) over waterproofing membrane that extends to the bridge deck and approaching roadway (Figure A.5). In this case, an expansion device is not employed. This detail, with 2-2

27 minimum length for approach slab of 14 ft, should not be used when the bridge length exceeds 250 ft, on bridges with integral abutments, nor when the approach roadway is concrete. 2. Concrete approach slab overlaid by 3 hot bituminous pavement over waterproofing membrane that extends to the bridge deck (Figure A.4.). In this case, an expansion device is employed. In this and the next case, the minimum length of approach slab is 20 ft (length revised on February 29, 1999). 3. Concrete approach slab with Bridge Expansion Device. This is used primarily with bare concrete bridge deck and adjacent concrete roadways (Figure A.3). When the adjacent roadway is concrete, an expansion device is required between the end of the roadway and end of approach slab. In all cases, the approach slab is anchored to the abutment. The length of concrete approach slab ranges from 14 ft to 30 ft with a constant thickness of 12 inches. All of CDOT concrete approach slabs have a sleeper slab (discussed later) placed directly on the abutment backfill. Dr. Trever Wang from CDOT Bridge Staff suggested that the length of approach slab be related to the depth of abutment of the bridge. Note that a longer approach slab requires a structural design with possibly larger concrete thickness and amount of reinforcements. 2.4 Abutment Backfill Conventional Granular Class 1 Structure Backfill Before 1992, CDOT measures to alleviate the bridge bump problem included the extension of wing walls along the roadway shoulder, and use of approach slab and granular backfill (Class I Structural Backfill) behind the abutments. The materials and construction requirements for Class-1 Backfill are presented in Table 2.1. Lift thickness is limited to 6 inches before compaction. 2-3

28 Table 2.1. CDOT Material and Construction Requirements for the Granular Class 1 Backfill. Requirements 1. Gradation 50 mm, (% Passing) 100 Sieve # 4 ((% Passing) Sieve # 50 (% Passing) Sieve # 200 (% Passing) Liquid Limit (%) <35 3. Plasticity Index (%) <6 4. Dry Unit Weight (kn/m 3 ) 95% of AASHTO T Flowfill Structural Backfill In November 1992, a CDOT Bridge Structural Worksheet was created for the use of flowfill (a low-strength concrete mix) backfill behind the abutment wall. The material requirements for the flowfill backfill are listed in Table 2.2. The current details for flowfill approaches are shown in Figure A.1. This revision was introduced to reduce shallow approach settlements (those resulting from the backfill), to prevent softening/erosion of the fill even if water infiltrated through the joints, and to improve constructibality/compactability of the fill behind the walls and around corners. Table 2.2. CDOT Material Requirements for Flowfill Backfill. Ingredient Lbs/C.Y Cement 50 Water 325 (or as needed) Coarse Aggregate 1700 (AASHTO N. 57 or 67) Fine Aggregate (AASHTO 1845 M 6) The maximum lift thickness for flowfill is 3 feet and placement of additional layers is not permitted until the flowfill has lost sufficient moisture to be walked on without indenting more than 2 inches. Vibration to consolidate flowfill as in regular concrete is not required in CDOT construction specifications for flowfill. Some CDOT construction engineers argued that vibration 2-4

29 of the flowfill would further help alleviate the bridge bump problem (vibration-induced settlements of flow fill up to 2 were reported) and would stiffen the flowfill and allow it to set faster. Before May, 31, 2000, the fill was placed on 1:1 slope (not the 2:1 slope as shown in Figure A.1). The construction of a 1:1 embankment slope with adequate compaction is difficult. The 2:1 slope also provides a smoother transition in stiffness from the abutment backfill to the adjacent embankment, further reducing the potential for approach settlement. This change in the slope, however, increased the quantities for flowfill especially in structures with deep abutments Mechanically Stabilized Class-1 Backfill Details for MSE abutment Class-1 backfill instead of flowfill were introduced in CDOT on May 21, 2000 (Figure A.2). Most of the reinforcements in the MSE embankments are geofabric wrapped around at the abutment, but geogrid (stiffer) are considered in some situations to stiffen the backfill and reduce settlements. The reinforced fill behind the abutment is used to build a vertical, self-contained wall capable of holding an approximately vertical shape and forming an air gap between the abutment and retained fill (Figure A.2). Thermal expansion and contraction of the bridge superstructure (decks and girders) cause lateral displacement of the approach backfill. This is a more critical factor with the use of integral abutment bridges, where abutment walls are strongly attached to the superstructure without joints. To overcome this problem, a system was developed where a very small gap (around 15 cm) is incorporated between the reinforced fill and the bridge abutment (Reid et al. 1998). It is hypothesized that the gap behind the abutment would allow for the thermally-induced movements of the integral abutment without affecting the backfill, thus reducing the applied passive stresses on the backfill soil to near zero. At the same time, this system would help to mobilize the shear strength of the retained approach fill and tensile resistance of the reinforcement, thus reducing the horizontal active soil pressure on the abutment wall. In later CDOT details, this gap was filled with 3 thick compressible low-density polystyrene or collapsible cardboard (Figure A.2) and it is also employed with flowfill approaches (Figure 2-5